Key Takeaways
Extended Takeaways
- Cannabis exposure was common at admission: 43% were current users, 67% of current cannabis users used daily, and 13% of the full sample had multiple substance use, supporting routine substance-use assessment in psychiatric inpatient care for young adults with bipolar disorder and psychotic disorders.
- Cannabis use alone did not clearly predict rehospitalization across the full year after discharge after adjustment for covariates (HR=1.6; 95% CI, 1.0–2.5), so clinicians should interpret cannabis use as part of a broader early-risk profile rather than as a sustained 12-month predictor by itself.
- Family and illness-course factors carried independent risk: absent or conflicted family relationship was associated with readmission (HR =2.2; 95% CI, 1.4–3.2), and psychiatric hospitalization history was similarly associated (HR =2.1; 95% CI, 1.4–3.2).
- The observed readmission burden was substantial even outside cannabis status, with 95 of 187 patients readmitted within 1 year and a median delay of 95 days (interquartile range 1–3: 15; 349), which argues for follow-up planning well before the typical 3-month mark.
- Specialized addiction care appeared underused relative to need: 37% had a SUD diagnosis, but only 12% received specialized care for addiction, and among those readmitted within the first month, current specialized addiction treatment was 6.8% versus 12.8% in all others.
- Objective toxicology was infrequent despite high reported drug exposure, with toxic urinary screening performed for only 30% of patients; when young adults with bipolar disorder or psychotic disorders present with deterioration, selective toxicologic testing may help clarify substance-related contributors.
